[SI-LIST] Re: Series termination value
- From: Bill Dempsey <BDempsey@xxxxxxxxxxxx>
- To: "Si-List (E-mail)" <si-list@xxxxxxxxxxxxx>
- Date: Fri, 22 Mar 2002 18:59:00 -0600
Justin,
You should really re-read Scott's and Lou's email and then look at yours.
They are clearly pointing out the flaws of your analysis in a nice manner.
This is *transient* analysis -- rise times matter. Imagine if it's a one
time pulse with an edge rate of <1ns. Into a 4", 6", 10" line you're going
to need some sort of series term. Applying a DC analysis will get you
burned...
Happy SI-ing.
Bill
p.s. There are classes available in SI if you're interested
Hi Louis;
Speed of light in a vacuum is 11.8 billion inches per secound- frequency is
90 MHz - without a calculator it is about 1200 inches. He mention only that
his frequency is 90 MHz, of course if there is more to it then that then it
changes the whole complextion of the problem but as the problem was stated
it sounded pretty straight forward. By the way didn't we meet ?
Justin
-----Original Message-----
From: Sanchez, Louis
Sent: Friday, March 22, 2002 2:05 PM
To: Tabatchnick, Justin
Subject: RE: [SI-LIST] Re: Series termination value
Hi Justin..........
I'm not sure how you arrived at the wave length. I may have missed part of
the email thread on this subject, but I didn't see either data rates, edge
rates, or line lengths mentioned in the original email message. Did I miss
something? If I did, please let me know. I believe that these parameters are
germain to the formation of a problem statement.
I think that the problem can first be broken down according to whether it's
a DC situation, or that the data and edge rates contain high frequency
spectral components that cause PCB traces to look like transmission lines.
If the edge rates are on the order of a nano-second or less, as is common in
our SONET applications (I realize the application was not specified), then a
reasonable estimate of the velocity of propogation in FR-4 material with a
71 ohm impedance is about 5.6 inches per nano-second.
Let me present an example for our discussion: assume; edge rate=1Ns, trace
length=12 inches, Vp=5.6 inches per nS. According to these conditions, the
round trip delay of this line is about 4.3 nS. Since the edge rate of the
pulse is 1 nS, the pulse doesn't know what is at the end of the line for 4.3
ns. Therefore, the driver will deliver a pulse of current into the line
equal to the voltage amplitude divided by the sum of driver source impedance
and transmission line impedance.
This pulse of current will continue to be delivered until the first
reflection comes back to the source 4.3 nS later. At that time the pulse of
current will drop to zero (assuming the load impedance is very very high).
When the trailing edge of the pulse occurs, the same thing will happen
except that because this is occuring on the trailing edge of the pulse, the
pulse of current delivered into the transmission line will be negative
going. This behaviour under the stated conditions can be observed with
SigExplorer (in Spectraquest) using the voltage and current probe features.
Best Regards,
Lou A. Sanchez
LOC3-14
Optical Product Group
Intel Sacramento
-----Original Message-----
From: Tabatchnick, Justin [mailto:justin.tabatchnick@xxxxxxxxx]
Sent: Friday, March 22, 2002 1:04 PM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: Series termination value
I have to jump in here but from the looks of things this is just a standard
transmission line problem. Impedance seen by the driver is a function of the
load impedance, the line characteristic impedance , the distance between the
impedance reference point ( in this case the 20 ohm resistor) and the load
impedance, and the wave length. The wave length in this case is about 1000
inches ( in air ) and since your trace is probably less then a foot , I
would say that you can treat this as a DC problem with the total load
impedance seen by the driver as 91 ohms. I don't see a problem with
attenuation due to loss or mismatch.
Justin Tabatchnick
SI Group
Intel, Sacramento
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List archives are viewable at:
//www.freelists.org/archives/si-list
or at our remote archives:
http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
Other related posts:
